Boxing Clever?: Piaget showed that 10-month-old infants will persist in looking for a toy in box A, where it has been placed several times, even after having been shown that it has been moved to box B, whereas 12-month-old infants do not. This phenomenon marks a developmental milestone in human infant cognition that Topál et al. (p. 1269 ; see the Perspective by Tomasello and Kaminski and the news story by Pennisi ) explored in a remarkable series of comparative tests. The results support the view that infants and adult dogs will both persevere in searching erroneously in box A because they regard the placement of the toy by a human experimenter as a social teaching event. By contrast, wolves rapidly learn correctly to search box B. They also observed that infants are able to generalize and thus still persevere when one experimenter places the toy in box A and a second then places the toy in box B. Dogs, however, display episodic learning, and a second experimenter reduces their searching choice to chance.

Having repeatedly retrieved an object from a location, human infants tend to search the same place even when they observe the object being hidden at another location. This perseverative error is usually explained by infants' inability to inhibit a previously rewarded search response or to recall the new location. We show that the tendency to commit this error is substantially reduced (from 81 to 41%) when the object is hidden in front of 10-month-old infants without the experimenter using the communicative cues that normally accompany object hiding in this task. We suggest that this improvement is due to an interpretive bias that normally helps infants learn from demonstrations but misleads them in the context of a hiding game. Our finding provides an alternative theoretical perspective on the nature of infants' perseverative search errors.

We tested whether dogs and 14–16-month-old infants are able to integrate intersensory information when presented with conspecific and heterospecific faces and vocalisations. The looking behaviour of dogs and infants was recorded with a non-invasive eye-tracking technique while they were concurrently presented with a dog and a female human portrait accompanied with acoustic stimuli of female human speech and a dog’s bark. Dogs showed evidence of both con- and heterospecific intermodal matching, while infants’ looking preferences indicated effective auditory–visual matching only when presented with the audio and visual stimuli of the non-conspecifics. The results of the present study provided further evidence that domestic dogs and human infants have similar socio-cognitive skills and highlighted the importance of comparative examinations on intermodal perception.

The comments by Fiset and Marshall-Pescini et al . raise important methodological issues and propose alternative accounts for our finding of perseverative search errors in dogs. Not denying that attentional processes and local enhancement are involved in such object search tasks, we provide here new evidence and argue that dogs’ behavior is affected by a combination of factors, including specific susceptibility to human communicative signals.